I have an architecture that houses many individual instances of splunk. Occasionally splunk-optimizes will run at the same time for many instances. This causes load spikes. Is there any way to stagger or control when splunk optimizes are run on a per instance basis.
You can alter the number of concurrent instances of splunk-optimize by changing the value set for maxConcurrentOptimizes
in indexes.conf but i dont think you can tell him when to run
outside forcing it to run via ./splunk-optimize <directory>
You can alter the number of concurrent instances of splunk-optimize by changing the value set for maxConcurrentOptimizes
in indexes.conf but i dont think you can tell him when to run
outside forcing it to run via ./splunk-optimize <directory>
you might want to open a case to splunk support to see if they have a suggestion
I'm not sure maxConcurrentOptimizes will help in my situation since its single optimizes spanning multiple instances. It seems that a possible ugly solution is to just disable it and cron them staggered but that would take some R&D. I was hoping there was a way to do it natively.